Document 13387382

advertisement
Submitted for publication (Paper id: 225)
INCOSE International Symposium, 20-23 June 2011, Denver CO, U.S.A
Satellites to Supply Chains, Energy to Finance —
SLIM for Model-Based Systems Engineering
Part 1: Motivation and Concept of SLIM
Copyright © 2011 InterCAX LLC and Georgia Institute of Technology . Published and used by INCOSE with permission.
Manas Bajaj1* , Dirk Zwemer1, Russell Peak2, Alex Phung1, Andrew Scott1, Miyako Wilson2
1. InterCAX LLC
75 5th Street, Suite 213,
Atlanta GA 30308 USA
www.intercax.com
2. Georgia Institute of Technology
Model-Based Systems Engineering Center,
MARC, 813 Ferst Drive, Atlanta GA 30318 USA
www.mbse.gatech.edu
Abstract
Development of complex systems is a collaborative effort spanning disciplines, teams,
processes, software tools, and modeling formalisms. Increasing system complexity, reduction in
available resources, globalized and competitive supply chains, and volatile market forces
necessitate that a unified model-based systems engineering environment replace ad-hoc,
document-centric and point-to-point environments in organizations developing complex systems.
To address this challenge, we envision SLIM—a collaborative, model-based systems
engineering workspace for realizing next-generation complex systems. SLIM uses SysML to
represent the front-end conceptual abstraction of a system that can “co-evolve” with the
underlying fine-grained connections to models in discipline-specific tools and standards. With
SLIM, system engineers can drive automated requirements verification, system simulations,
trade studies and optimization, risk analyses, design reviews, system verification and validation,
and other key systems engineering tasks from the earliest stages of development directly from
the SysML-based system model. SLIM provides analysis tools that are independent of any
systems engineering methodology, and integration tools that connect SysML with a wide variety
of COTS and in-house design and simulation tools.
We are presenting SLIM and its applications in two papers. In Part 1 (this paper), we present
the motivation and challenges that led to SLIM. We describe the conceptual architecture (section
1) and use cases (section 2) of SLIM followed by tools available for production and evaluation
usage (section 3). In Part 2 paper—SLIM Applications—we present the applications of SLIM
tools to a variety of domains, both in traditional as well as non-traditional domains of systems
engineering. Representative examples from space, energy, infrastructure, manufacturing and
supply chain, military operations, and bank systems are presented.
*
Corresponding author: Manas Bajaj, manas.bajaj@intercax.com, phone: +1-404-592-6897. Preferred citation:
Bajaj, M., Zwemer, D., Peak, R., Phung, A., Scott, A. and Wilson, M. (2011). Satellites to Supply Chains, Energy to Finance
— SLIM for Model-Based Systems Engineering, Part 1: Motivation and Concept of SLIM. 21st Annual INCOSE International
Symposium, Denver, CO, June 20-23, 2011.
1
1 Introduction to SLIM (Systems Lifecycle Management)
SLIM is an integrated software platform for systems lifecycle management. It is envisioned
to provide capabilities that combine the strengths of model-based systems engineering and
product lifecycle management (PLM). Though the scope of SLIM is from systems design to
delivery and sustainment, this paper primarily focuses on systems design and analysis aspects of
SLIM and related tools available to date. In this section, we first present the motivation for SLIM
in terms of gaps in current state-of-the-art commercial tools for design and analysis of complex
systems. Following this, we present a conceptual architecture of SLIM and describe key end-user
capabilities that SLIM provides.
1.1 Motivation
Figure 1 illustrates two types of fundamental gaps in currently available tools for systems
design and analysis. These gaps are described below in the context of system lifecycle phases,
and are generally applicable to both traditional and non-traditional systems engineering domains.
As an example, Figure 1 shows the lifecycle phases for typical NASA systems (NASA 2007).
Pre-Phase A: Concept Studies
Gap 2a
Phase B: Preliminary Design &
Technology Completion
Phase D: System Assembly,
Integration and Test, Launch
Phase E: Operations &
Sustainment
Phase F: Closeout
Time
Project Life-Cycle Phases
[NASA SE Handbook, 2007]
Phase A: Concept & Technology
Development
Phase C: Final Design and
Fabrication
Conceptual
system
analyses
Conceptual
system
design
Gap 1
Lack of continuity of
system MoEs
(performance, cost,
risk,…)
Gap 2b
Detailed
system
analyses
Detailed
system
design
System Design Maturity
Figure 1: Gaps in current state-of-the-art tools for design and analysis of complex systems
Gap 1 represents lack of model-based continuity of system engineering activities from the
early phases (proposals and conceptual design) to detailed phases (detailed design, development
and delivery). Gap 1 exists because the tools used for systems modeling and analyses are
different in each phase. In several domains, conceptual design phases largely depend on the use
of diagramming tools and spreadsheets. Tools used in a design phase capture only those aspects
of the system that are relevant for that phase, and hence there is no continuity of system
definition, requirements, performance parameters, and their evaluation results from one phase to
another. Results of design exploration, uncertainty in design parameters, assessment of systemic
risk, and other cross-phase parameters and studies are typically buried in mammoth and difficultto-maintain spreadsheets, documents, and proprietary domain-specific tools thereby making it
difficult to track and communicate them effectively during phase transitions.
2
Information forwarding for core aspects of the system, such as the requirements and
objectives, structure and behavior, performance and risk parameters, cost, power and weight
budgets, and its development process (tasks, resources and milestones) need to be maintained
continuously through the system design and development phases independent of the diverse
COTS tools used for representing and managing these aspects. In this light, some of the
questions posed by systems engineering and project management teams are:
•
How does one ensure that models defined in tools during early phases and later phases are
representations of the same system?
•
How does one ensure that the requirements and performance parameters evaluated in tools
during early phases and later phases concern the same system or even the same project?
•
How does one relate the analysis and evaluation results obtained during early phases and
those obtained during later phases?
•
How does one propagate trade study results, uncertainties in system parameters, and systemic
risk assessment from the preliminary design phases to detailed design phases, and eventually
to system deployment and sustainment?
Gap 2 represents disconnects between the concerns in a given phase, such as disconnects
between design and analysis/simulation models in a design phase. Gap 2 manifests in
heterogeneous model transformations beyond design and analysis, such as between requirements
and structure, logical structure and physical structure, and structure and behavior. Typically,
individual modeling and simulation tools represent only specific aspects of a system, such as
requirements, structure, and behavior, and do not provide a holistic model of the entire system.
As a result, systems engineering activities can at best be realized using point-to-point data flows,
custom model transformations, and workflow automation software to keep all aspects in sync.
There is an urgent need in systems engineering for using a single coherent system model that can
federate domain-specific design and analysis models in a configuration managed environment.
1.2 Conceptual architecture
SLIM is a software environment for integrated model-based systems engineering. As shown
in Figure 2, it uses SysML as the front-end for multi-disciplinary teams to collaboratively
develop a unified, coherent representation of the system from the earliest stages of development.
The system model (in SysML) can ‘co-evolve’ with the associated domain-specific models, such
as Computer-Aided Design and Engineering (CAD/CAE) models. Relationships between the
system model and the domain-specific models can range from qualitative dependency relations
to quantitative acausal parametric relations which are executable on-demand for seamless model
traceability and interoperability. The SysML-based system model is a conceptual abstraction of
the system that has sufficient details for orchestrating all systems engineering activities, ranging
from automated requirements verification and system trades to risk analysis and system V&V.
This unified system model is not a data store but a map of the system which can be used to
federate domain-specific models of different aspects of the system.
3
Primavera, MS Project, Excel,…
Project Management
DE
CAD
DE
DE
Requirements
SLIM
NX, Pro/E, CATIA, …
DOORS, RequisitePro,…
SysML
DE
DE
x
SE
SE
z=f(x,y)
Optimization
z
Simulation/CAE
y
MATLAB, Simulink, NPSS, ABAQUS, ANSYS, SINDA/FLUINT, pSPICE, Mathematica,…
ModelCenter, Isight,…
else
X > x1
SE
DE
DE
Libraries / Databases
PLM System
Teamcenter, Windchill,…
DE: Domain expert
SE: System engineer
DE
Manufacturing, Supply Chain
CAD models, cost models, analysis modules, parts and material databases, supplier database, …
Tecnomatix, SAP, …
Figure 2: Conceptual Architecture of SLIM
SLIM builds on existing SysML authoring environments, such as MagicDraw (No Magic
2010), Artisan Studio (Atego 2010), Rhapsody (IBM 2010), and Enterprise Architect (IBM
2010), and PLM environments, such as Teamcenter (Siemens PLM 2010) and Windchill (PTC
2010). SLIM provides plugins to SysML authoring environment for facilitating systems
engineering design, integration, and verification and validation (V&V) flows directly from the
SysML model. Automated requirements verification, system trades and optimization, risk
analysis, sizing, cost and performance estimation, producibility analyses, model-based design
reviews, and more can be triggered from the SysML model. At its core, SLIM provides software
tools that are independent of any systems engineering methodology, and are building blocks for
supporting different types of design and verification flows. We envision that SLIM will provide
system engineering teams the flexibility to define, test, and deploy organization-specific
workflows using the building block tools.
1.3 SLIM as a design environment
SLIM provides a rich and seamless design environment for complex systems engineering,
which includes organizations, humans, operating environment, and other aspects in addition to
the system-of-interest. Figure 3 illustrates the difference between existing design environments,
generalized for multiple domains, and SLIM’s design environment for realizing complex
4
systems. Typical design environments for complex systems are driven by document-based
integration or point-to-point model-based integration between models representing different
aspects of the system. Lack of specialized tools for conceptual design, unlike CAD and CAE
tools for detailed design, leads to usage of diagramming tools (such as PowerPoint (Microsoft
2010) and Visio (Microsoft 2010)) with spreadsheets for numerical analysis. Workflow
automation tools provide a means to transfer data from one tool to another at best. A significant
amount of time and resources are spent by system engineering teams in the administration of
islands of automation and document-based interfaces, and producing reports for system design
and project reviews.
SLIM’s Design Environment
Existing Design Environment
Project Management
DE
Requirements
DE
SE
DE
SE
CAD
SE
DE
Simulation/CAE
Libraries / Databases
DE
DE
Optimization
Manufacturing, Supply Chain
DE
DE: Domain expert
SE: System engineer
PLM System
DE
Reviews, Meetings, Administration
Figure 3: Comparison of existing design environment and SLIM’s design environment for complex systems
In contrast, SLIM’s design environment is driven by a SysML-based system model that is the
centerpiece of the design process from the beginning. SLIM provides tools that aid system
engineering activities directly from the SysML-based system model. SysML structure and
behavior constructs (such as blocks and activities) replace drawing tools, while parametrics
replaces spreadsheets. Automated unit checking, requirements verification, and trades can now
be performed from the SysML model. Changes in the design architecture automatically reflect in
the analysis models unlike manual updates from drawing tools to spreadsheets.
Since no single modeling language can represent all aspects of a system throughout its
lifecycle, SLIM provides tools that integrate SysML with a wide array of in-house and COTS
design and simulation tools, and documents and spreadsheets for data presentation and reporting.
The level of integration can range from qualitative dependency relationships for traceability,
interface specifications for model generation and reconfiguration, and/or quantitative
relationships for model synchronization. Coupled with enterprise PLM systems, SLIM enables
change and configuration management, conflict resolution, and model-based communication
between stakeholders.
SLIM provides libraries of reusable models—SysML constructs connected to native design
and simulation models—that make it easy to assemble and perform optimization and trades on
5
the system model. Examples of libraries for aircraft systems include design objects (wing,
fuselages, tails, and airfoils), analysis models (cost, sizing), system behavior (mission profiles),
materials, and manufacturing constructs. Examples of libraries for supply chains include libraries
of: (a) suppliers, (b) customers, (c) parts and assemblies, (d) costing models, and (e) supplier
allocation and optimization models.
For complex systems, SLIM provides a powerful alternative to “PowerPoint and Spreadsheet
engineering” and point-to-point integration of analysis codes (spreadsheets, MATLAB/Simulink,
C++, Fortran, Java) disintegrated from the system design definition.
1.4 Preliminary design and evaluation of advanced concepts
SLIM directly answers the following challenge—the conceptual design process has the
highest impact on the product lifecycle but the least availability of design and decision support
tools. SysML, the front-end of SLIM, is a general-purpose systems modeling language and an
open standard that is not confined to any specific design phase or methodology. SLIM serves as
a decision support environment for systems-of-systems. The system-of-interest can be codeveloped with the related operating environment, organization, humans, manufacturing
facilities, government regulations, and market behavior. With a unified SysML model as the
centerpiece, the impact of changes in requirements or any other element can be studied on both
qualitative and quantitative levels.
SLIM provides analysis tools (section 3) for executing SysML constructs representing
different aspects of the system, such as SysML parametric models for cost, performance,
reliability, and SysML activity models for mission profiles. As a result, systems engineers can
perform system trades, risk analyses, and automated requirements verification on system
alternatives on a continuous basis. SLIM’s integration tools (section 3) will allow system
engineers to connect the SysML-based system model to design and analysis models in COTS
tools—concept sketches in CAD tools and orbit planning models in STK (AGI 2010)—for
driving system analyses during conceptual phases.
A key factor in the incorporation of new designs, materials, and manufacturing technologies
into new systems is the ability to model the entire end-to-end system lifecycle. SysML can
incorporate issues of producibility, maintainability and disposal into the system model from the
earliest stages of conceptual design. While SysML provides rich and extensible constructs for
modeling “systems-of-systems”, SLIM enables qualitative and quantitative “what-if” analyses
and trades for both unconventional and conventional systems built on combinations of new and
existing technology.
1.5 Transition from conceptual design to detailed design
SLIM provides plug-n-play and automated model re-configuration capabilities to swap lowfidelity analysis models with high-fidelity ones, while still maintaining the conceptual
relationships (in SysML) between the models. The suite of integration tools in SLIM (section 3)
allows system engineers to wrap externally-defined design and simulation models and codes as
SysML constructs and replace (or combine) low-fidelity by high-fidelity analyses. Bi-directional
6
relationships between the SysML-based system model and domain-specific models in COTS
tools, as shown in Figure 2, represent both qualitative and quantitative relationships that enable a
wide range of model interoperability services—traceability, transformation, reconfiguration,
synchronization, and conflict resolution. Design, behavior, and requirements-related parameters
in the SysML model can link to spreadsheets during the conceptual design phases or CAD and
CAE models during the detailed design phases (Figure 5).
1.6 Detailed design and analysis
SLIM builds on several years of design-analysis integration research (Peak, Paredis et al.
2005; Peak, Burkhart et al. 2007; Peak, Burkhart et al. 2007; Bajaj 2008) that provides patterns
and methods to automatically compose SysML-based simulation templates—parametric models
relating design models to analysis models—from a library of analytical building blocks for
variable topology systems. These patterns make it possible to abstract analysis model
formulation from solution, thereby allowing designers and analysts to rapidly (re)formulate
analysis models and simulation templates from high-level specifications and also explore
different solution strategies (e.g. FEM, FDM, Meshless) and solvers (e.g. ANSYS, ABAQUS)
for an analysis problem. Trades and sensitivity analyses can be automatically triggered from the
SysML-based simulation templates that also provide specifications for design space exploration
and multi-disciplinary design optimization using tools such as ModelCenter (Phoenix Integration
2010) and Isight (Dassault Systèmes SIMULIA 2010).
SLIM’s approach for relating requirements, design, and analysis does not require each
engineer to learn the others’ tools or compromise his or her own models. Figure 4 shows the
general concept. Requirements, design, and analysis concerns can be modeled for every
component in the system model. Component-level requirements, derived from decomposing
system-level requirements, are pushed from the system model to specialists (such as designers
and analysts) while design parameters and analysis results flow back into the system model.
Parametric relationships can exist between concerns (design, analysis, requirements) for a given
component, and across the system hierarchy (such as for weight and cost roll-ups).
Systems Engineering Domain (SysML)
Design Domain
(e.g. CAD, STK)
System Model
Component Z
System Model
Property A
Property B
Figure 4: Interaction between systems
engineer, designer, and analyst.
B = a+b
Component Z
CAD Model
Component Z
CAD Design
Property a
Property b
Parameter a
Parameter b
Figure 5: Later-stage systems engineering – design - analysis
integration.
SLIM provides multiple types of model-connectivity patterns that can be used to connect the
SysML-based system model to externally-defined design and analysis models. System engineers,
7
designers, and analysts can select the connectivity mode best suited for the purpose. Both filebased exchange (STEP, XML) and live model-based interaction (via native API) are supported.
Figure 5 above illustrates one of these patterns used for relating detailed design models to the
SysML model. The CAD design for Component Z resident in the CAD tool is mirrored
automatically by SLIM in the SysML modeling environment. The CAD model parameters,
however, may not correspond exactly to the system model properties needed to evaluate other
aspects of the system. For example, a CAD model may calculate true wing area, while systems
calculations require the wing reference area. SysML parametrics is used to connect the CAD
model parameters and the system model parameters at a fine-grained level. Execution of
parametric relationships enables bi-direction information flow.
2 SLIM Use Cases
The goal of SLIM is to facilitate model-based design, analysis, review, change management,
procurement, and sustainment of complex systems. Figure 6 is a SysML use case diagram that
illustrates some of the key systems engineering use cases that SLIM will address. The use cases
are described in this section in an anti-clockwise order of appearance in the figure, starting with
the use case Facilitate model-based design and analysis.
Figure 6: Key systems engineering use cases of SLIM
SLIM tools that are currently available or in-development are primarily focused on modelbased design and analysis of complex systems. This includes tools for parametric analysis,
orchestrating system simulations, performing trade studies and risk analyses, and automated
verification of requirements.
8
SLIM will facilitate model-based design reviews. Design reviews in existing design
environments involve manual creation of an extensive set of documents that have to be carefully
scanned by review teams. In SLIM’s design environment, reviewers will be able to invoke all
system alternatives that were analyzed, the types of analyses performed, the analysis results, and
design decisions and rationale directly from the SysML-based abstraction of the system.
Document-based artifacts can be automatically generated from the SysML model for archival
and future reference.
A unified system model can represent not only the design and analysis aspects of a system
but also test procedures necessary for verification and validation (V&V) and their relationships
to system and sub-system requirements. SLIM can facilitate model-based V&V directly from
the SysML model. Test activities can be scheduled and automatically executed each time a new
version of the system model is checked into a repository. This capability is similar to unit and
functional testing and project automation commonly used in software development projects, but
for large systems engineering projects.
SLIM can facilitate model-based procurement and delivery of complex systems. The
SysML-based system model is not only useful for facilitating the development of complex
systems but also as a contractual artifact binding the relationship between an organization and its
customers, suppliers, and collaborators. Customers can issue request-for-proposals (RFPs) using
requirements, use cases, and functional specifications defined using SysML. Proposals and
deliverables in the form of SysML-based design models can then be automatically verified and
analyzed against specifications. After a system has been delivered and deployed, the SysMLbased system model serves as a blueprint of system structure, behavior, and target performance
rating, much like design drawings and access plans for aircrafts and ships. System upgrade,
maintenance, repair, and other sustainment-related plans can then be mapped and generated
from the system model.
Design environments for collaborative development of complex system need to provide
configuration control and version management capabilities. SLIM will leverage project
management and configuration control capabilities native to commercial-strength PLM tools
such as Teamcenter.
With the existence of a system model, SLIM can facilitate model-based change
management. Changes in customer requirements, available resources, system design and
behavior are common in large systems engineering projects. With SLIM, system engineers can
analyze the impact of changes via visualization and parametric analyses from the SysML model.
Such analysis will be crucial to planning, controlling, and affecting changes in the system design
process.
SLIM will provide tools for creating libraries of reusable system constructs, such as those
based on structural, parametrics, and behavior concepts. Examples for aircraft systems include
libraries of aircraft wings and fuselages, libraries of aircraft maneuvers and mission profiles, and
libraries of cost, weight, and sizing analysis models. Elements in the library can be defined
natively in SysML, or can be SysML elements wrapping externally-defined library elements
such as CAD and STK models.
9
3 SLIM Components
In this section, we present SLIM tools that are available for production and evaluation usage.
Figure 7 is a SysML block definition diagram that shows the conceptual decomposition of the
SLIM framework. SLIM tools are accessible and driven from SysML authoring environments,
and will access externally-defined models in PLM environments in the future. The main
components of the SLIM framework are listed below, followed by a detailed description of tools
and their availability.
Figure 7: Conceptual decomposition of the SLIM framework
ƒ SysML analysis tools provide capabilities to trigger system-level analyses and automated
requirements verification from a SysML model. This includes (1) Parametric solver for
executing parametric relationships, (2) Trade study tool for performing trade studies using
parametric models, (3) Risk analysis tool for performing Monte Carlo simulations based on
parametric and activity models, and (4) Activity execution tool for executing SysML activity
models.
ƒ SysML visualization tools provide capabilities to visualize SysML models, in addition to the
nine (9) types of SysML diagrams, for enhanced model traceability. Flattened and hierarchical
graph- and table-based views can better highlight cross-cutting relationships.
10
ƒ SysML integration tools provide capabilities to integrate the SysML-based system model with
models and data based on other open standards (such as STEP) or defined in COTS and inhouse tools.
ƒ SysML-based libraries provide building blocks for creating and executing the SysML-based
system model. Library elements could be defined natively in the SysML environment, such as
library of math functions for parametric analyses and trade studies, or involve SysML elements
wrapping CAD, CAE, and STK models in PLM environments.
In sections 1-2, we presented the vision and use cases of SLIM. In this section, we present SLIM
tools and capabilities that are: (1) available commercially for production usage, (2) available as
beta for early adopters, and (3) are under development. We invite users to participate in this
effort by using existing tools, providing feedback, and contributing to the development of new
tools under the SLIM framework. Two categories of SysML-based tools are presented here.
These are: (1) SysML analysis tools, and (2) SysML integration tools.
3.1 SysML analysis tools
3.1.1 Parametric solver
The Parametric solver tool executes SysML parametric models using math solvers such as
Mathematica (Wolfram Research 2010), OpenModelica (OSMC 2010), and MATLAB (The
MathWorks 2010). Parametrics is one of the four pillars of SysML and provides modeling
constructs to define acausal, fine-grained, mathematical relationships between SysML block
properties. Parametrics is applicable to a broad range of systems engineering use cases. For
example, parametric relations could represent the following scenarios:
•
Design constraints, such as sum of the weights of all parts should be equal or less than a predefined value.
•
Simulation templates that relate design and analysis models, such as the calculation of cost
variables, key performance parameters (KPPs), risk metrics, and other measures of
effectiveness (MoEs) from the design model.
•
Mathematical relationships between different design views and analysis views, such as
relationships between properties of logical and physical system structure, between properties
of hardware and software components, or between SysML blocks representing physical
system components and the corresponding CAD models.
•
Property-based requirements—refinement of text-based requirements—represented as
mathematical constraints that can then be automatically verified for design alternatives.
Due to its broad applicability, SLIM’s Parametric solver is a powerful systems engineering
tool. It can be used for verifying or solving design constraints; computing system MoEs, KPPs,
and risk metrics; synchronizing diverse views of the system; automatically verify requirements;
and connecting, synchronizing, and controlling externally-defined models (CAD/CAE/Simulink)
with SysML models.
11
Figure 8 shows a parametric model (in MagicDraw) of the FireSat system model being
developed by the INCOSE SSWG Challenge Team (INCOSE). This parametric model relates
performance metrics (ground coverage and scan resolution) and annual operational cost of the
FireSat to the design and operational variables, such as orbit altitude, sensor resolution, and
angle covered by the optical instruments. The performance metrics and the cost variable are
highlighted in red in the parametric diagram. The parametric model also shows constraints used
for verifying requirements. The ResolutionReqt constraint block is used for the rR1 constraint
property (bottom right corner of the parametric diagram) and is a mathematical representation of
a text-based requirement concerning FireSat scan resolution. It compares the scan resolution to
the required value (30m) and computes the ScanResVerify property that represents the result of
requirement verification (1 for pass and 0 for fail).
Figure 8: SysML parametric model for computing coverage, scan resolution, and flight cost of the FireSat,
and for verifying resolution requirement
Figure 9 shows the trade study results generated by ParaMagic® (InterCAX - ParaMagic
2010)—SLIM’s Parametric solver for MagicDraw. Each row represents a trade study scenario
(given values of design and operational variables) for 2 FireSats. In this case, ParaMagic® read
input values for each scenario from the spreadsheet; computed target variables (value properties
flightCost, coveragePerDay, and scanResolution in Figure 8); verified them against
requirements; and exported results back to the spreadsheet. The columns Cost req., Coverage
req., and Res. req. correspond to the results of verifying cost, coverage, and resolution (for each
FireSat) requirements—1 for pass and 0 for fail. The results indicate that when the satellites
12
operate at lower orbits, the annual costs exceed the operating budget and the coverage is not
sufficient to meet the requirements. On the other hand, when the satellites operate at higher
orbits, the resolution requirement fails. Rows highlighted in green are scenarios for which all
requirements were successfully met. Sensitivity plots and other graphs can also be generated by
the Parametric solver.
Satellite 1
Satelite 2
Altitude Ang. Aperture Altitude Ang. Aperture
km
deg
km
deg
300
3
300
3
325
3
325
3
350
3
350
3
3
375
3
375
3
400
3
400
425
3
425
3
450
3
450
3
475
3
475
3
500
3
500
3
525
3
525
3
550
3
550
3
575
3
575
3
3
600
3
600
Satellite 1
Satellite 2
Cost/yr.
Cost req. Coverage/day Coverage req. Tgt. Resolution Res. req. Tgt. Resolution Res. req.
M$/yr
1‐pass, 0‐fail M sq km/day 1‐pass, 0‐fail
meters
1‐pass, 0‐fail
meters
1‐pass, 0‐fail
77.23
0
2.50
0
15.71
1
15.71
1
51.61
0
2.70
0
17.02
1
17.02
1
36.28
0
2.89
0
18.33
1
18.33
1
26.65
0
3.08
1
19.64
1
19.64
1
20.36
0
3.26
1
20.94
1
20.94
1
19.67
1
3.45
1
22.25
1
22.25
1
19.67
1
3.63
1
23.56
1
23.56
1
19.67
1
3.81
1
24.87
1
24.87
1
19.67
1
3.99
1
26.18
1
26.18
1
19.67
1
4.17
1
27.49
1
27.49
1
19.67
1
4.34
1
28.80
1
28.80
1
19.67
1
4.52
1
30.11
0
30.11
0
19.67
1
4.69
1
31.42
0
31.42
0
®
Figure 9: Trade study and requirement verification results generated and reported by ParaMagic
Figure 10 shows a parametric diagram in Artisan Studio where the SimulinkHomeHeating
constraint block wraps a MATLAB script M-file used for executing a Simulink model.
ParaSolver (InterCAX - ParaSolver 2011) —SLIM’s Parametric solver for Artisan Studio—
executes all relationships in the parametric model, including the wrapped Simulink model, to
compute the target properties. In a similar manner, constraint blocks can wrap user-defined
Mathematica functions. Figure 10 also shows the ParaSolver browser (bottom left) which
presents the parametric models for a system alternative in an object-oriented spreadsheet
structure. When a user selects a specific system component, the parametric relations owned by
that component can be seen in the relationship table of the browser.
SLIM’s Parametric solver can be used by system engineers to orchestrate system simulations
and automate analysis workflows. System engineers can create parametric models that chain
externally-defined MATLAB functions or scripts, Mathematica functions, Simulink models to
compute key performance parameters and other MOEs.
SLIM’s SysML Parametric solver is available for production usage for the following SysML
authoring tools:
•
•
•
•
ParaMagic® (InterCAX - ParaMagic 2010)for MagicDraw
Melody™ (InterCAX - Melody 2010) for Rational Rhapsody
ParaSolver™ (InterCAX - ParaSolver 2011) for Artisan Studio
Solvea™ (InterCAX - Solvea 2011) for Enterprise Architect
These tools are being used for system design and analysis at leading organizations in aerospace,
defense, supply chain, energy, and electronics and telecommunications sectors.
13
par Home
connected to a Simulink model
SHH : SimulinkHomeHeating
constraints
{cost=xfwExternal(matlab,scriptascii, demoscriptasciisimulink,row,col,outtemp,daycyc)}
outtemp : Real
OD_Home : Outdoors
Temp : Real
daycyc : Real
DailyCycle : Real
row : Real
OutputRow : Real
col : Real
cost : Real
OutputColumn : Real
DailyCost : Real
ParaSolver for Artisan Studio
Simulink model executed from SysML
Figure 10: Parametric constraints wrapping Simulink models executed by ParaSolver (for Artisan Studio)
3.1.2 Trade study tool
The Trade study tool provides system engineers the capability to setup and execute trade
studies directly from the SysML model. In the current version, system engineers can execute
parametric models in a batch mode for a set of scenarios that have been specified explicitly or
implicitly as intervals or discrete values.
Figure 13 illustrates the configuration view of the Trade study tool. The configuration view
shows the SysML instance model in the left hand pane that serves as a fixed topology template
for parameter variations in the trade study. Users can identify inputs, constants, and outputs in
the trade study, and specify the data sources and targets for the input and output variables
respectively. Inputs can be setup in an explicit manner by specifying discrete values or
connecting to a cell range in an Excel spreadsheet, or in an implicit manner by specifying a range
with increments. The Trade study tool automatically generates scenarios (if specified implicitly)
by permuting all combinations of input variables. While running a trade study, the tool shows the
specific scenario for which the parametric models are being executed and the specific solver
being used. After execution, the Trade study tool exports results to the spreadsheets configured
during setup. Post-processing and plotting capabilities of Excel can be used for displaying and
reporting trade study results.
14
The Trade study tool is commercially available with each of the Parametric solver tools—
ParaMagic®, Melody™, ParaSolver™, Solvea™.
Figure 11: Trade study tool – configuration view
In the future, the Trade study tool will be enhanced to perform design exploration during system
trades, potentially leveraging optimization and design exploration capabilities of ModelCenter
and Isight.
3.2 Risk analysis tool
SLIM’s Risk analysis tool will provide system engineers the capability to compute system
MoEs, KPPs, cost, and other metrics given the uncertainties in input variables—design,
operation, project, and others. Given the probability distributions for input variables, the Risk
analysis tool will run Monte Carlo-type simulations on SysML parametric models and/or activity
models to calculate probability distributions and related statistical metrics for output variables
(such as system MoEs and KPPs). Such analyses directly benefit system engineers developing
proposals or designing complex systems in performing trades between performance, cost,
schedule, and risk.
Figure 12 shows a SysML parametric model for a military operation where a defense system
composed of aircrafts, UAVs, ground forces, and analysis cells is trying to track and destroy
time-sensitive targets (TSTs). The parametric model (shown in Rational Rhapsody) is used to
compute: (1) the response time of the defense system once a time-sensitive target is spotted, (2)
the probability that the target is able to fire a missile before being destroyed, and (3) the
probability that the target is eventually destroyed before it goes into hiding. The spreadsheet in
the figure shows the input and output variables (in red) for this analysis. The output variables
indicate if the missile was fired by the time-sensitive target (TST), if the TST was destroyed, and
the response time of the overall defense system. A value of 1 indicates success and a value of 0
15
indicates failure. Input values were generated based on normal probability distributions with
given mean and variance. For a defense system alternative (specified number of ground troops,
aircrafts, and cells), the analysis shows 36% mean probability that a missile is fired by the target
and 100% probability that the target is destroyed before it goes into hiding. The response time
for the scenarios is shown using a histogram and a cumulative frequency curve. This analysis
was performed using Melody™—SLIM’s Parametric solver tool for Rhapsody.
The Risk analysis tool builds on capabilities available with the Trade study tool, and is
available with all Parametric solver tools—ParaMagic®, Melody™, ParaSolver™, and Solvea™.
par [ block] Miss ion [Mission_PAR]
1
1
itsGlobalHawk
Effec tiveDetectionTime:Real
1
itsJCC
1
itsAnalysisCe ll
EffectiveFlightTime:Real
t2:Real
t1:Real
itsSpecialForces
EffectiveTravelTime:Real
itsF18
StdAnalysisTime:Real
1
1
StdAssignmentTime:Real
t3:Real
t4:Real
t5:Real
Mission.Resp1:Response
Constraints
t = t1 + t2 + t3 + max(t4, t5)
t:Real
1
itsTST
ResponseTime
MinFiringTime:Real
MaxFiringTime:Real
tfmax: Real
tfmin: Real
rt:Real
1
«Constrain tPrope rty»
Mi ssion.PMF1:ProbMF
Constraints
pmf = if(rt > tfmax,1,0)
pmf:Real
TargetDestroyed
1
rt:Real
MissileFired
«Constrain tProperty»
ptd:Real
Mi ssion.PTD1:ProbTD
tfmin:Real
tfmax:Real
Constraints
ptd = if(rt > (tfmin + tfmax),0,1)
Figure 12: Computing success probabilities & response times for military operations using Melody™ for Rhapsody
3.3 Activity execution tool
SysML activity models are used for modeling function-based behavior of systems, such as
for modeling mission profiles for multi-mission space systems, or financial and business
processes. Execution of activity models provides system engineers and project managers the
ability to simulate and analyze system behavior. Figure 13 illustrates a SysML activity model for
driving mini-rovers (as shown in the picture). Execution of this activity model generates and
executes python code that governs the movement of the rover. The example also demonstrates
the use of SysML for system operations. The same SysML model used for designing the system
can be used for operating it. This prototype capability is available as a plugin called MyroMagic
(Georgia Tech - MyroMagic 2011) for MagicDraw.
16
Figure 13: Activity model for driving mini-rovers using the MyroMagic plugin
3.4 SysML integration tools
3.4.1 SysML-Excel integration
The SysML-Excel integration tool allows system engineers to connect Excel spreadsheets to
SysML models. SysML is used for modeling the system architecture, behavior, and related
parametrics, while Excel is used for data storage, presentation, and reporting. With the SysMLExcel integration tool, system engineers can connect system properties to cells in spreadsheets,
and automatically read from or write to Excel. Figure 14 shows the Excel Setup utility for the
SysML-Excel integration tool. In the Excel Setup utility, properties configured to read from
Excel are shown in blue and those configured to write to Excel are shown in red. After setup,
users can invoke Excel read/write operations from the SysML model. This integration capability
is available with all Parametric solver tools—ParaMagic®, Melody™, ParaSolver™, and
Solvea™.
17
Figure 14: SysML-Excel integration tool - ParaMagic®, Melody™, ParaSolver™, Solvea™
3.4.2 SysML-MATLAB/Simulink integration tool
The SysML-MATLAB/Simulink integration tool allows users to wrap MATLAB M-files
(functions or scripts) as SysML constraint blocks and reuse them in SysML parametric models.
During execution, the Parametric solver tool invokes MATLAB/Simulink for executing
constraint relationships wrapping M-files and brings results back to SysML or uses it for solving
next set of constraint relationships. This capability is illustrated in Figure 10 and is available
with all Parametric solver tools—ParaMagic®, Melody™, ParaSolver™, and Solvea™.
3.4.3 SysML-Mathematica integration tool
The SysML-Mathematica integration tool allows users to wrap user-defined Mathematica
functions (.m files) as SysML constraint blocks and reuse them in SysML parametric models, in
the same manner as MATLAB M-files. In addition, the Parametric solver tool uses Mathematica
as the core solver for executing parametric relationships. This capability is available with all
Parametric solver tools— ParaMagic®, Melody™, ParaSolver™, and Solvea™.
3.4.4 SysML-OpenModelica integration
The Parametric solver tool can also use OpenModelica (free) as a core solver for executing
parametric relationships. This capability is available with all three Parametric solver tools—
ParaMagic®, Melody™, ParaSolver™, and Solvea™. In the future, SLIM will support the use of
SysML-Modelica libraries based on the SysML-Modelica transformation specifications (Object
Management Group 2010).
18
3.4.5 SysML-CAD (MCAD/ECAD) integration
The SysML-CAD integration tool provides system engineers the ability to connect SysML
models to CAD models. This capability is necessary for: (a) integrating CAD-based concept
sketches with the system model during preliminary design, (b) transitioning from conceptual to
detailed design, and (c) maintaining a consistent view of the system and its components during
detailed design, especially for cases where multiple CAD models (MCAD, ECAD) are being
concurrently developed for a single component.
Figure 15 illustrates a generalized approach for integrating SysML and design models (e.g.
CAD, STK). The integration approach is shown for a single sub-system component that needs to
be connected to its detailed definition in the design tool (CAD, STK). Step 1 shows the SysMLbased system model (LHS) as being developed by the system engineer, and the detailed design
model (RHS) as being developed by a designer. The CAD integration tool automatically
generates a view of the CAD model in the SysML modeling environment along with the
mapping relationships, as shown in Step 2. Then, the system engineer and the designer can relate
the properties of the CAD model view and the system model using SysML parametrics, as shown
in Step 3. SysML parametrics is a key to model-based communication between domains. In this
case, the parametric relationships represent knowledge that is often lost in documents,
spreadsheets, and other ad-hoc modes of communication. Capturing this knowledge in the
SysML model makes it traceable. Execution of parametric relationships, using the Parametric
solver tool, achieves bi-directional flow of information between the system model and the CAD
model, as shown in Step 4. System engineers can visualize and compute dependencies between
the system model and domain-specific design models, and automatically verify system
requirements on a continuous basis.
Step 1
Systems Engineering Domain
Step 2
Design Domain
System Model
Component Z
System Model
Component Z
CAD Design
Property A
Property B
Parameter a
Parameter b
Step 3
Step 4
Systems Engineering Domain
Design Domain
System Model
Data flow automated by Parametric solvers
Component Z
System Model
Property A
Property B
B = a+b
Component Z
CAD Model
Component Z
CAD Design
Property a
Property b
Parameter a
Parameter b
Figure 15: SysML–Design (MCAD/ECAD, STK) integration capability – end-user steps
19
Figure 16 and Figure 17 illustrate the SysML-CAD integration tool for a Mini-Satellite
system. One of the components (BGA) of the Mini-Satellite is being designed in NX (Siemens
PLM 2010) CAD tool. The system engineer wants to connect the parameters of the BGA
component in the system model to the properties of its detailed CAD definition in order verify
component and system-level requirements. The SysML-CAD integration tool automatically
generates a SysML view of the CAD model, as shown in Figure 17 (RHS). Then, the system
engineer creates parametric relationships between the BGA component in the system model
(LHS in Figure 17) and the SysML view of its CAD definition (RHS in Figure 17). The
parametric relationships are used to compute the bounding box and the weight of the BGA
electronic component from its CAD definition. The Parametric solver tool is used to execute
these and other parametric relationships to achieve information flow between the CAD model
and the SysML-based system model. Changes made to the BGA CAD model in NX can
automatically be reflected in the BGA component of the Mini-Satellite and vice versa. This
interaction is based on a live connection between the SysML authoring environment (e.g.
MagicDraw) and NX, which is established by SLIM’s SysML-CAD integration tool. With this
capability, a system engineer can automatically pull the latest updates from the detailed design
model into the SysML model, and use this information for performing system-level analyses,
verifying requirements, performing trades, and other system V&V tasks.
System model in SysML
CAD model in Siemens NX
Figure 16: Mini-Satellite system model in SysML. The BGA electronic component and its CAD
definition are highlighted.
20
Systems Engineering Domain
Design Domain
System Model
Data flow automated by Parametric solvers
Component Z
System Model
Component Z
CAD Model
Component Z
CAD Design
Property A
Property B
Property a
Property b
Parameter a
Parameter b
B = a+b
Figure 17: Parametric relationships between the BGA component in the system model and its
CAD definition.
The SysML-CAD integration tool will be deployed as Shape™ plugin (InterCAX - Shape
2011) for different SysML authoring environments. Currently, the Shape™ plugin is available for
MagicDraw SysML tool and NX CAD tool at a beta-level maturity.
3.4.6 SysML-STK integration
The SysML-STK integration tool works on the same integration pattern as shown in Figure
15. It allows system engineers developing aerospace systems to integrate the SysML-based
system model with orbit planning and analysis models in STK. SysML parametrics is used to
achieve fine-grained synchronization between the SysML-based system model and the STK
model—a capability similar to that shown for SysML-CAD models in Figure 17 above.
Figure 18 illustrates the SysML-STK integration plugin (called Apogee™) in action for the
FireSat model. Here, Apogee™ is used for sending system-level design parameters from the
SysML model to the STK / AGI Component simulation environment, and retrieving STK
simulation results back into the SysML environment for requirement verification and further
processing. For example, the sensor descriptions of the FireSat are sent to the STK environment
and the chain and coverage response results are retrieved from STK simulations. The chain
response provides information on the specific time period during which the FireSat has access to
specific features-of-interest on the ground. The coverage response provides information on how
21
much of a specific feature-of-interest (absolute and relative area) would be accessible by the
FireSat in a given time interval.
Currently, the Apogee™ plugin (InterCAX - Apogee 2011) is available for MagicDraw
SysML tool at a demonstrator-level maturity.
Chain analysis
Coverage analysis
FireSat model (STK)
g
Figure 18: Apogee™ plugin for integrating SysML and STK models
3.5 SysML-based libraries
The Parametric solver tool comes with a library of constraint blocks that wrap commonly
used math functions, such as trigonometric, logarithmic, exponential, hyperbolic, conditionals
(if-else), and aggregate functions. Complex functions can be composed from simpler functions
using SysML parametrics. Constraint blocks can also wrap externally-defined functions and
scripts, such as MATLAB function and script M-files and Mathematica functions.
The SysML-CAD and -STK integration tools make it possible to reflect libraries of CAD
parts and STK design/analysis elements in the SysML environment, which can then be used for
composing system models. Libraries of system-level analysis models, such as cost models,
performance models, and reliability models can be developed directly using SysML constructs,
such as parametrics and activity models) or as SysML constructs wrapping externally-defined
models, such as library of constraint blocks wrapping MATLAB routines. In the future, SLIM
22
will leverage SysML-Modelica transformation specifications (Object Management Group 2010)
to make libraries of analytical models in Modelica accessible from the SysML modeling
environment.
3.6 SysML visualization tools
3.6.1 Panorama
Panorama (Georgia Tech - Panorama 2011) is a SysML-based visualization tool that flattens
parametric models for a given system alternative. Figure 19 below illustrates flattened SysML
parametric models for a supply chain and for an electronics recycling network.
supply chain
electronics recycling network
Figure 19: SysML parametric models (flattened) in Panorama for supply chain and electronics recycling network
4 Summary & Future Work
This paper (Part 1) presents our vision of SLIM as a collaborative model-based systems
engineering workspace that uses SysML as the front-end for orchestrating system engineering
activities from the earliest stages of system development. The SysML-based system model
serves as a unified, conceptual abstraction of the system independent of the specific design and
analysis tools that shall be used in the development process. SLIM provides analysis tools to
compute system MoEs and KPPs, orchestrate simulations, perform trades, analyze risk,
automatically verify requirements, and perform other systems engineering activities from
SysML. It also provides plugins to integrate the system model (in SysML) to a variety of design
and analysis tools, such as CAD tools, STK, CAE/FEA tools, Simulink, and math solvers—
MATLAB, Mathematica, OpenModelica, and Excel. SLIM enables a consolidated design
environment where the SysML-based system model is used for model-based X, where X =
design, analysis, reviews, V&V, deployment, and procurement. SLIM tools and capabilities are
driven from a SysML authoring environment, and work with models that are configuration
controlled in the enterprise PLM environment.
23
Several key components of the SLIM framework are available for production usage, as
described in section 3. Other important capabilities are under-development. In section 2, we have
presented the key systems engineering use cases that define the roadmap for SLIM. We invite
interested users and organizations to help shape this roadmap, use existing SLIM tools and
provide feedback, and contribute to the development of SLIM.
In Part 2 paper—SLIM Applications—we present applications of SLIM in aerospace, energy,
infrastructure, manufacturing and supply chain, and bank system domains.
References
AGI (2010). "STK." Retrieved Oct 24, 2010, from http://www.stk.com/.
Atego (2010). "Artisan Studio." Retrieved Oct 24, 2010, from
http://www.atego.com/products/artisan-studio/.
Bajaj, M. (2008). Knowledge Composition Methodology for Effective Analysis Problem
Formulation in Simulation-based Design. G.W.Woodruff School of Mechanical Engineering,
PhD, Georgia Institute of Technology; PhD Dissertation: http://hdl.handle.net/1853/26639.
Dassault Systèmes SIMULIA (2010). "Isight." Retrieved Oct 24, 2010, from
http://www.simulia.com/products/isight.html.
Georgia Tech - MyroMagic (2011). "MyroMagic Plugin for Mobile Rovers." Retrieved Mar 28,
2011, from http://www.buzztoys.gatech.edu/buzztoys/index.html.
Georgia Tech - Panorama (2011). "Panorama Plugin for MagicDraw." Retrieved Mar 28, 2011,
from http://www.buzztoys.gatech.edu/buzztoys/index.html.
IBM (2010). "Rational Rhapsody." Retrieved Oct 24, 2010, from http://www01.ibm.com/software/rational/products/rhapsody/designer/.
INCOSE. "Space Systems Working Group (SSWG) and Challenge Team." 2010, from
http://www.omgwiki.org/MBSE
http://www.incose.org/practice/techactivities/wg/sswg/.
InterCAX - Apogee (2011). "Apogee plugin for SysML- STK / AGI Components integration."
Retrieved Mar 30, 2011, from www.intercax.com/sysml.
InterCAX - Melody (2010). "Melody - SysML Parametric Solver and Integrator for Rational
Rhapsody." Retrieved Oct 24, 2010, from http://www.intercax.com/melody.
InterCAX - ParaMagic (2010). "ParaMagic® - SysML Parametric Solver and Integrator for
MagicDraw." Retrieved Oct 24, 2010, from http://www.intercax.com/sysml
http://www.magicdraw.com/paramagic.
24
InterCAX - ParaSolver (2011). "Artisan Studio ParaSolver ". Retrieved Jan 31, 2011, from
www.atego.com/products/artisan-studio-parasolver/
www.intercax.com/sysml.
InterCAX - Shape (2011). "Shape plugin for MagicDraw." Retrieved Mar 30, 2011, from
www.intercax.com/sysml.
InterCAX - Solvea (2011). "SolveaTM - SysML Parametric Solver and Integrator for
MagicDraw." Retrieved Mar 28, 2011, from www.intercax.com/solvea.
Microsoft (2010). "PowerPoint." Retrieved Oct 24, 2010, from http://office.microsoft.com/enus/powerpoint/.
Microsoft (2010). "Visio." Retrieved Oct 24, 2010, from http://office.microsoft.com/en-us/visio/.
NASA (2007). Systems Engineering Handbook, NASA/SP-2007-6105 Rev 1. Washington, D.C.
20546.
No Magic (2010). "MagicDraw SysML modeling tool, version 16.9." Retrieved Oct 24, 2010,
from http://www.magicdraw.com/sysml.
Object Management Group (2010). "SysML and Modelica Integration." Retrieved Oct 24, 2010,
from http://www.omgwiki.org/OMGSysML/doku.php?id=sysmlmodelica:sysml_and_modelica_integration.
OSMC (2010). "OpenModelica 1.5." Retrieved Oct 24, 2010, from www.openmodelica.org.
Peak, R., Paredis, C. J. J., Tamburini, D. R., Bajaj, M., Kim, I. and Wilson, M. (2005). The
Composable Object (COB) Knowledge Representation: Enabling Advanced Collaborative
Engineering Environments (CEEs), COB Requirements & Objectives (v1.0). The Georgia
Institute of Technology, Oct 31, 2005. http://www.eislab.gatech.edu/projects/nasangcobs/COB_Requirements_v1.0.pdf
Peak, R. S., Burkhart, R. M., Friedenthal, S. A., Wilson, M. W., Bajaj, M. and Kim, I. (2007).
Simulation-Based Design Using SysML Part 1: A Parametrics Primer. The Seventeenth
International Symposium of the International Council on Systems Engineering, San Diego,
California, USA, June 24 -28, 2007. http://eislab.gatech.edu/pubs/conferences/2007-incose-is-1peak-primer/2007-incose-is-1-peak-primer.pdf.
Peak, R. S., Burkhart, R. M., Friedenthal, S. A., Wilson, M. W., Bajaj, M. and Kim, I. (2007).
Simulation-Based Design Using SysML Part 2: Celebrating Diversity by Example. The
Seventeenth International Symposium of the International Council on Systems Engineering, San
Diego, California, USA, June 24 -28, 2007. http://eislab.gatech.edu/pubs/conferences/2007incose-is-1-peak-primer/2007-incose-is-1-peak-primer.pdf.
25
Phoenix Integration (2010). "ModelCenter." Retrieved Oct 24, 2010, from http://www.phoenixint.com/software/phx_modelcenter.php.
PTC (2010). "Windchill." Retrieved Oct 24, 2010, from http://www.ptc.com/products/windchill/.
Siemens PLM (2010). "NX." Retrieved Oct 24, 2010, from
http://www.plm.automation.siemens.com/en_us/products/nx/.
Siemens PLM (2010). "Teamcenter." Retrieved Oct 24, 2010, from
http://www.plm.automation.siemens.com/en_us/products/teamcenter/.
The MathWorks (2010). "MATLAB." Retrieved Oct 24, 2010, from
http://www.mathworks.com/products/matlab/.
Wolfram Research (2010). "Mathematica." Retrieved Oct 24, 2010, from
http://www.wolfram.com/products/mathematica/index.html.
Biography
Manas Bajaj, PhD is the Chief Systems Officer at InterCAX (www.InterCAX.com). He has successfully
led several government and industry-sponsored projects, including prestigious SBIR awards from NIST
and NASA. Dr. Bajaj’s research interests are in the realm of SysML and model-based systems
engineering (MBSE), computer-aided design and engineering (CAD/CAE), advanced modeling and
simulation methods, and open standards for product and systems lifecycle management (PLM/SLM). He
is the originator of the Knowledge Composition Methodology for simulation-based design of complex
variable topology systems. At INCOSE, he is a core team member of three MBSE Challenge Teams –
Space Systems, Modeling and Simulation, and Smart Grid. He has authored several publications and won
best paper awards. Dr. Bajaj earned his PhD (2008) and MS (2003) in Mechanical Engineering from the
Georgia Institute of Technology, and B.Tech. (2001) in Ocean Engineering and Naval Architecture from
the Indian Institute of Technology (IIT), Kharagpur, India. He has been actively involved in the
development, implementation, and deployment of the OMG SysML standard and the ISO STEP AP210
standard for electronics. He is a Content Developer (author) for the OMG Certified Systems Modeling
Professional (OCSMP) certification program, and coaches organizations on SysML and MBSE. Dr. Bajaj
is a member of INCOSE and a contributor in OMG and PDES Inc. working groups.
Dirk Zwemer, PhD is President and CEO of InterCAX (www.InterCAX.com). Dr. Zwemer has over
thirty years experience in the electronics industry with Bell Labs, Exxon, ITT, SRI Consulting and other
organizations. He is the author of three patents and multiple technical papers, trade journal articles, and
market research reports. Prior to joining InterCAX, he held positions as VP Technology, VP Operations,
and President of AkroMetrix LLC, a leader in mechanical test equipment and services for the global
electronics industry, where he was actively engaged in market channel development, advertising and
tradeshows, and licensing. He received a PhD in Chemical Physics from UC Berkeley and an MBA from
Santa Clara University. Dr. Zwemer provides strategic consulting for customers and shows how to apply
SysML in a wide variety of domains. He is certified systems modeling professional (OCSMP Model
Builder Advanced) and a member of the Smart Grid Challenge Team within the INCOSE MBSE
Initiative.
26
Russell Peak, PhD is a Senior Researcher at the Georgia Institute of Technology where he serves as
Director of the Modeling & Simulation Lab (www.msl.gatech.edu) and Associate Director of the ModelBased Systems Engineering Center (MBSE-C) (www.mbse.gatech.edu). Dr. Peak specializes in
knowledge-based methods for modeling & simulation, standards-based product lifecycle management
(PLM) frameworks, and knowledge representations that enable complex system interoperability. Dr. Peak
originated the multi-representation architecture (MRA)—a collection of patterns for CAD-CAE
interoperability—and composable objects (COBs)—a non-causal object-oriented knowledge
representation. Dr. Peak leads the INCOSE MBSE Challenge Team for Modeling & Simulation
Interoperability † with applications to mechatronics (including mobile robotics testbeds) as a
representative complex systems domain. He is a Content Developer (author) for the OMG Certified
Systems Modeling Professional (OCSMP) program, and coaches organizations on SysML and MBSE. Dr.
Peak earned his Ph.D., M.S., and B.S. in Mechanical Engineering from Georgia Institute of Technology.
Alex Phung is a senior research engineer at InterCAX LLC. He has 12 years of experience in a wide
range of software development technologies. Mr. Phung is a certified systems modeling professional
(OCSMP Model Builder Intermediate) and has over five years of experience in UML and SysML
software development. Mr. Phung earned his M.S. and B.S. in Mechanical Engineering from Texas A&M
University.
Andrew Scott is a research engineer at InterCAX LLC. He is an experienced SysML modeler and
software developer. Mr. Scott earned his B.S. in Mechanical Engineering at Georgia Institute of
Technology.
Miyako Wilson is a research engineer at the Manufacturing Research Center at Georgia Institute of
Technology. She specializes in object-oriented software development. Ms. Wilson earned her M.S. in
Mechanical Engineering at Georgia Institute of Technology and B.S. in Mechanical Engineering at
Arkansas State University.
†
http://www.omgwiki.org/MBSE/doku.php?id=mbse:modsim
27
Download